Traditional experiments (like A/B tests) often assign treatments randomly and stick to the same plan throughout. However, adaptive experimental design learns as it goes; it updates how treatments are assigned based on the data collected thus far. This gathers clearer answers with fewer resources.
On 22 May, NERA will host the Economists in Tech 2025 session “Adaptive Experimental Design for Efficient Average Treatment Effect Estimation and Treatment Choice.” In this seminar, Mizuho-DL Financial Technology Financial Engineer Masahiro Kato will introduce adaptive experimental design—a modern approach that helps organizations and researchers make better decisions when testing new ideas, treatments, and policies. The session will be facilitated in NERA’s Tokyo office by Managing Director Hans Ihle.
During this session, Dr. Kato will focus on:
Estimating impact – How to measure the average effect of a treatment (e.g., does a new drug work better than the old one?) more accurately and efficiently.
Choosing the best option – How to quickly and reliably figure out which treatment or policy works best among several choices.
